Jiaxin Fang

Publication

BridgeWAM overview: latent bridge queries connect video and action experts, with simulation and real-world evaluation results.

2026 · Preprint

BridgeWAM: Bridging Video and Action Experts via Latent Bridge Queries

Jiaxin Fang, Ruitong Liu, Zeyu Lai, Xingyu Wang, Mingyuan Jia, Pingyue Sheng, Weikang Wan, Mengdi Xu, Yuxiang Gao

Submitted to ICLR 2027 Main Track

arXiv:2610.14746

ViGAR overview: visual subgoal planning and subgoal execution for compositional and in-context robotic manipulation.

2026 · Preprint

Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation

Shukai Gong, Xuanran Zhai, Yintianrun Zhang, Ruopeng Cui, Ye Huang, Yiyang Fu, Dexuan Lyu, Chaojie Li, Xinyi Song, Peiwen Lin, Chuang Wang, Mingyuan Jia, Yufan Deng, Jiaxin Fang, Bo Liang, Jiaxin Li, Yuxiang Gao, Hao Liu, Daquan Zhou

arXiv:2610.02368

DFM-VLA architecture overview: language, vision, and noisy action tokens enter a bidirectional transformer for iterative action refinement.

2026 · Preprint

DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching

Jiayi Chen, Wenxuan Song, Jiaxin Fang, Ruiqing Yin, Jingbo Wang, Shuai Chen, Jieyuan Pei, Yikai Qin, Feifan Chen, Haodong Yan, Zhide Zhong, Wen Chen, Yan Wang, Yuxiang Gao, Haoang Li

arXiv:2603.26320v4

FeMo overview: momentum-based feature prediction across diffusion timesteps and a comparison with TaylorSeer in feature space.

FeMo qualitative comparisons with other acceleration methods on FLUX and DiT image generation.

2025 · Preprint

Accelerate Diffusion Transformers with Feature Momentum

Jiaxin Fang, Chang Zou, Jiacheng Liu, Yuanhuiyi Lyu, Xuming Hu, Linfeng Zhang

OpenReview · PDF